Wikipedia:Reference desk/Archives/Computing/2012 September 8

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

September 8

i want tcs current details

hi ! iam going to attend tcs india interview.i want tcs current details.like its achievements,clients,services,employees ,revenue etc.i tried in net.but i am getting old information.can any one provide? — Preceding unsigned comment added by Phanihup (talk • contribs) 01:11, 8 September 2012 (UTC)

Care to tell us which TCS from that link you mean, if any ? Tata Consultancy Services ? StuRat (talk) 01:45, 8 September 2012 (UTC)

Which is the actual benefit of a Tesla card in rendering?

We have a very little little render farm (actually it’s just nine core i7 connected by LAN) and we render mostly in ‘mental ray’, ‘scanline’, and ‘quicksilver’ (all autodesk 3ds max 2012) Each pc consists of an i7-2600 + an Asrock p67pro3 + 16 gb ram + GTS450 (with quadro2000 drivers) + etc… and all linked by LAN Having in mind the rendering motors that I use, my question is… how exactly a Tesla would affect my rendering times? Sure I could buy one and then give it a try, but these things are too much expensive, I mean real expensive… that’s why I’m asking. Other thing… someone knows if there is the possibility of transform a GeForce card in a Tesla equivalent?? (Kind like I did with the gts450-> quadro2000 but for rendering instead of modeling proposes) Iskánder Vigoa Pérez 05:11, 8 September 2012 (UTC) — Preceding unsigned comment added by Iskander HFC (talk • contribs)

If you are rendering as you are, a Tesla would make absolutely no difference to your render times using mental ray and the scanline renderer. Quicksilver renders would be accelerated by the Tesla units, and you could expect a considerable speed increase (although compiling the shaders will still be done on the CPU). Unless you are using a renderer that does GPU rendering (iray, Vray-RT, Octane, etc) the Tesla GPU will not be used. Sadly, mental ray does not support GPU rendering yet.

If you are using mental ray, you would be better off using DBR (distributed bucket rendering) for stills, and backburner to schedule animation renders.

One thing to remember with GPU rendering is that the whole scene + textures must fit in the memory of the GPU card/unit.

I hope this helps Zzubnik (talk) 13:45, 10 September 2012 (UTC)

If the quicksilver is the only one that “maybe” will be improved it seems I will pass because I only use it for computing the ambient occlusion, and it’s already very quick.

Using DBR + Bburner is what I do. Almost for now… I don’t like my results vs time with iray, and the octane render it’s just not apt enough, I have not seen this “Vray-RT”. It’s suitable for animations?

Thanks for your answer — Preceding unsigned comment added by Iskander HFC (talk • contribs) 14:18, 10 September 2012 (UTC)

You're welcome. Vray-RT has been used for animations. I just did a Youtube search and found a few things. I don't know that much about Vray, so I can't really answer that question. Zzubnik (talk) 09:14, 11 September 2012 (UTC)

Rotate Google Map?

I'm looking at a Google map where the street names I need to read run vertically. Is there an easy way to rotate the Google map 90 degrees to put the names upright? (While remaining in the Google Map program)

Thanks. Wanderer57 (talk) 12:46, 8 September 2012 (UTC)

In the Google labs, there is an add-on for Google maps that lets you rotate the maps. Zzubnik (talk) 13:47, 10 September 2012 (UTC)

Sagemath in LaTex

If you have something like:

\begin{sagesilent}
G=Graphics()
H=Graphics()
var('x')
G+=plot(cos(x),(x,0,pi))
H+= text("$\cos(x)$", (1, 1), fontsize=16, color='black')
\end{sagesilent}
\[ \sageplot{G+H}\] 

into a tex file, how can you tell Latex to place it on the left or right, and to make it smaller, or even to show it vertically? Note: the whole thing is showing correctly, but just too big in the middle of the page Qpl87 (talk) 12:36, 8 September 2012 (UTC)

Turing test for human scoring

I'd eventually like to get myself copied into an AI, but I'm concerned about whether I -- or any sufficiently accurate copy of me -- could pass a Turing test. How can I get myself scored on a Turing test, preferably against both computer and human norms? 69.158.1.219 (talk) 13:28, 8 September 2012 (UTC)

There isn't really "scoring" on Turing tests. It's not like a standardized assessment. It's just about whether other people can tell you from an AI bot or not. That's basically it. Do people routinely say, "wait, you're just a chatbot, aren't you?" If not, then you probably "pass". It's actually a very low bar, and there are chatbots that pass it routinely. I've met users on Wikipedia who I've found to be barely above a chatbot — they reply stiffly, they basically seem only capable of recycling existing content, they don't seem to be able to engage whatsoever — but it's a rare thing (and I'm unsure what accounts for it — language difficulties? autism or other spectrum disorder? general awkwardness? maybe they really are chatbots?). --Mr.98 (talk) 15:41, 8 September 2012 (UTC)

You could apply to be a human contestant in the Loebner Prize - a Turing Test based contest pitting humans against computers. Wired has an article written by a previous contestant - What's It Mean to Be Human, Anyway? - Cucumber Mike (talk) 20:13, 8 September 2012 (UTC)

Java

What are major and minor elements in JAVA 199 (talk) 14:11, 8 September 2012 (UTC)

I have a suspicion you can find the information you seek from Java (programming language). ¦ Reisio (talk) 15:15, 8 September 2012 (UTC)

No, this doesnt help, searched a lot. — Preceding unsigned comment added by Intr199 (talk • contribs) 16:11, 8 September 2012 (UTC)

Where did you see this terminology? Looie496 (talk) 17:10, 8 September 2012 (UTC)

Major and minor elements refer to the object model proposed by Grady Booch, not to Java specifically. According to Booch, for a language to be fully object-oriented it has to possess 7 elements, the major are abstraction, modularity, hierarchy, and encapsulation. The minor elements are persistence, typing, and concurrency. Programming languages may implement only some of these elements and be object-oriented to a different degree. ListCheck (talk) 17:41, 8 September 2012 (UTC)

What's the point of LaTex? II

I asked this question above, but now I want a different perspective. Mostly the answers' tendency was concentrated on what the user was doing (CVS, scripting, choosing what he means, but not what he sees). I want now to know from a typographical perspective. Is there any effect that can only be produced with a LaTex editors and not with Word/Quark X/Scribus/InDesign? It's clear that each editor has its limitations, I can't, for example, import pictures into my notepad .txt file, no matter how hard I try. 17:17, 8 September 2012 (UTC) — Preceding unsigned comment added by Comploose (talk • contribs) http://media-cache-ec6.pinterest.com/upload/59883870014903546_RaTehfZB.jpg

I think that depends on the definition of "can". In principle, Microsoft probably can implement formatting code that will yield typography comparable to (La)TeX. In practice, LaTeX produces better results. I once had to submit a CV in Word format, and, at first glance, thought it looked pretty good. Then I put it next to the LaTeX original, which quite changed my perception. I see two other advantages with LaTeX. First, I can simply type simple formulas, which is much faster than using the formula editor in Word (and, of course, the result looks better). Secondly, I can see the actual logical structure of the source and all formatting. In Word, I never know if that space between to italic words is, by accident, bold (or whatever ;-). --Stephan Schulz (talk) 17:29, 8 September 2012 (UTC)

In principle, anything LaTeX can do, other typesetting programs, markup languages, or word processors can also do. In practice, many people prefer LaTeX because they find it convenient. Other people find LaTeX inconvenient. It is strictly a matter of preference. Nimur (talk) 18:44, 8 September 2012 (UTC)

Is there any effect that can only be produced with a LaTex editors...? What do you mean by "LaTeX editors"? (La)TeX files are plain text files, so you can use any text editor including notepad to write your documents/texts without limitations. Also not sure what you mean by "effects". bamse (talk) 21:18, 8 September 2012 (UTC)

I suppose the OP mean some LaTex compiler. 80.31.74.137 (talk) 22:25, 8 September 2012 (UTC)

The standard LaTeX document styles tend to give a very "classic" sort of presentation. Serif fonts (except for beamer), extra space between sentences, lovingly crafted (almost calligraphic) symbols. To me this is preferable to the web-page-like stuff you get from Nimur's preferred HTML (at least, without a lot of special fiddling). But it's an aesthetic judgment. --Trovatore (talk) 22:02, 8 September 2012 (UTC)

You could also write a CSS that looks as classic as your LaTex template. That's not that common because people tend to use CSS and LaTex in different environments: the first in web-design and the second in typography design for printed material. And I do think is better to stick to each technology depending of your purpose (web or stuff to be printed (including pdfs).

Another difference to be considered is if you want to use the wysiwyg or the wysiwym paradigm. Only the last letter is different, but both approaches are miles apart. If you care about the appearance of your text, like in the case of a popular magazine, the first is more appropriate. If you care about the logical connection of the elements, like in the case of a scientific text, use the second. 80.31.74.137 (talk) 22:25, 8 September 2012 (UTC)

I disagree that you should use WYSIWYG if you care about the appearance. Appearance comes out better in the long run if you have a well-designed program that takes care of it automatically, based on the logical structure, than if you have to be fiddling with it all the time as you're writing. The only thing that's better about WYSIWYG is that it's easier to get something that looks approximately right, because it's a one-step process, whereas with something like LaTeX you likely have to make at least one extra pass to clean up stuff that the automatic approach didn't get quite how you want it.

It's quite imaginable that something similar to LaTeX could be implemented in CSS, though I don't know of any such package. I suspect, though, that it would be order-of-magnitude as big, complicated, and hard to learn, as LaTeX. --Trovatore (talk) 22:43, 8 September 2012 (UTC)

As I understand it, the CSS file is roughly equivalent to a LaTeX style file. The HTML is roughly equivalent to a LaTeX source file. And the rendering engine (e.g. Gecko or WebKit) is comparable to the TeX/LaTeX "compiler". The whole thing is slightly obscured because LaTeX is just a set of macros for TeX. --Stephan Schulz (talk) 23:13, 8 September 2012 (UTC)

Right, so what you'd need is some huge interconnected set of CSS files that can include each other. Or something. I really don't know much about CSS so I don't know how that works. But I'm sure you couldn't get equivalent functionality with a single CSS of reasonable size. --Trovatore (talk) 23:28, 8 September 2012 (UTC)

Yes, but no matter how good your CSS, unless your rendering engine supports the typographic intricacies that make TeX output look great, no amont of CSS will help. --Stephan Schulz (talk) 23:34, 8 September 2012 (UTC)

But which typographic intricacies are those? This is something Knuth fans always have trouble actually nailing down. Do you have a particular attachment to a specific interpolation-algorithm used by TeX to render some specific glyph? Anecdotally, most colleagues and readers I have experimented on can not tell the difference between a document prepared with TeX, and a document I have created in Microsoft Word and applied the Computer Modern font. Many years ago, TeX had a clear advantage over its competitors: the kerning and typesetting was more sophisticated. But today, every major text layout engine, whether intended for word-processing or web content, supports these features: they are taken as a given. While these features are still "intricate," I don't believe TeX's implementation is in any respect distinctive; this sort of support is what I consider basic functionality in every word-processor. Variable-width character glyphs are no longer novel. Advanced line-editing and layout is universally available. WYSIWYG editors, like LibreOffice, now support elaborate character editing - via both a graphical menu, or through direct markup; documents may embed styles or import style definitions from an external style sheet; so if the claim is that TeX offers anything special with respect to character layout and rendering, I call shenanigans. These are standard features. As far as I can tell, the chief advantage of using TeX is the ability to rapidly enter mathematical formulae, especially by skilled users who have memorized large quantities of style markup syntax. And personally, when I see Knuth's example of attention to detail... what I see is: 1) crudely color-matched paletted GIF images whose background color is neither transparent, nor identical to his HTML document; 2) significant aliasing due to a low-resolution render that has been upsampled; and 3) a discolored highlight surrounding the glyph. These artifacts affect both "bad-case" and "good-case" TeX renders. When somebody so famous for attention to detail can overlook these issues, I have very little patience when they proceed to complain about serif stylization. You can't have it both ways: fix these insignificant glyph errors, or don't worry about insignificant glyph errors. Nimur (talk) 00:22, 9 September 2012 (UTC)

Some people answering here have never worked with professional publishing programs like Scribus, QuarkXPress, or a handful of Adobe programs. They are fantastic for publishing and you don't have to constantly fiddle with anything. These programs are way better than Word, which seems what LaTeX fans have in mind when answering. At least in the Adobe versions your work is layered. That means that you can work on each layer (picture, headers, texts) independently, and in a team, different team members can concentrate on their layer. Can you do that with Kile, Lyx or similar programs? There's really no fancy 'effect' that you cannot put on paper using them, no scientific symbol that is not available. You can choose at what pixel you put any picture, any textbox. The main difference is still what I said above: constructing your work by what you see or by what you mean.

If you still don't believe me when I say that everything is possible, try Nimur's method above. Print a file, look at a printed file and try to guess if it was produced with LaTeX or with any Desktop publishing program.

Alternatively, download Scribus, which is free and give it a try. 79.155.47.134 (talk) 00:58, 9 September 2012 (UTC)

A very long time ago, I had a job typesetting journal articles that were often quite technical — lots of differential equations. All of this was in PageMaker (e.g. what InDesign was before it was InDesign). I could replicate each of the equations, but 1. they were not functional in any way (they were purely visual, with complicated kerning and line height settings to make them appear correctly), and 2. it was not easy, and took hours of fiddling to get correct. For the technically inclined, LaTeX is especially good at letting you creating functional equations (e.g. writing them out in a markup language that a computer could understand the purpose of) and simultaneously outputting those as visually correct. There are other alternatives to doing this sort of thing, but this is an area where LaTeX is superior over just using other, non-specialist word processors or page layout software or even CSS. They're just visual; they're not functional. There's a big gap there. --Mr.98 (talk) 04:17, 9 September 2012 (UTC)

"A very long time ago" could mean, I suppose, more than eight years, since PageMaker was discontinued. Since then, many things happen, and software developers indeed discovered that graphical tools have to deal with kerning, math symbols and math equations as such. Maybe you'll need an add-on like MathMagic, but I don't see anything typographical that cannot be done, unless you have LaTeX. OsmanRF34 (talk) 19:03, 9 September 2012 (UTC)

I remember "A very long time ago" (1986) using the AMX Pagemaker [1](what? no article?) on the BBC micro. Desktop Publishing has certainly changed since then! Dbfirs 08:25, 10 September 2012 (UTC)

One advantage of LaTeX is that, being a plain-text format, it doesn't need a specialized editor (great for programmers who do everything in Emacs already), and it is very amenable to machine-generated input. I find LaTeX's surface syntax painful, so I write in a different syntax and translate it to LaTeX. Writing the translator would've been too painful if LaTeX were a binary format instead of a language. Paul (Stansifer) 17:46, 10 September 2012 (UTC)

Reduction (complexity)

What is this sentence even means --> Intuitively, problem A is reducible to problem B if solutions to B exist and give solutions to A whenever A has solutions.? I can't understand what is it saying. Is that mean that the time to find solution to problem A and problem B are the same?Pendragon5 (talk) 22:06, 8 September 2012 (UTC)

No, not necessarily. It means that if you want to solve a problem a of type A (say a travelling salesman problem), you can construct a problem b of type B (say an instance of 3-SAT) and solve b instead. You can then convert the solution of b back into a solution of a. But the transformation from a into b and from the solution to b into a solution of a also take time (and that time can be very considerable). We are often interested in a polynomial-time reduction, where the time for the transformation is polynomial in the size of the problem description. For that kind of reduction, the problem remain in the same complexity class (for the major classes like P, NP, and NP-complete). --Stephan Schulz (talk) 22:49, 8 September 2012 (UTC)